Bug #107029 | deadlock when using trigger logic implementation | ||
---|---|---|---|
Submitted: | 14 Apr 2022 22:42 | Modified: | 15 Apr 2022 13:19 |
Reporter: | Александр Ммммммм | Email Updates: | |
Status: | Not a Bug | Impact on me: | |
Category: | MySQL Server: Locking | Severity: | S3 (Non-critical) |
Version: | 8.0.28 | OS: | Any |
Assigned to: | CPU Architecture: | Any |
[14 Apr 2022 22:42]
Александр Ммммммм
[15 Apr 2022 13:19]
MySQL Verification Team
Hi Mr. Mmmmmmm, Thank you for your bug report. However, it is not a bug. Deadlocks are expected behaviour for all pessimistic transactional engines. That is why application has to inspect all return calls and re-commit the transaction that is rolled back. If you have got a report on the deadlock from InnoDB storage engine, then it is a proof that a storage engine is working properly. There is a slight peculiarity in your report. It is about the lock escalation from the shared to the exclusive lock. That means that problem like yours could be solved in the release 8.0.29, where a feature request described in https://bugs.mysql.com/bug.php?id=21356, from this bug forum, has been implemented. It is possible that deadlocks like you described are solved in the release 8.0.29. Nobody yet knows when 8.0.29 release will be published. Not a bug.